Temperature-dependent regulation of flowering by antagonistic FLM variants

Journal name:
Nature
Volume:
503,
Pages:
414–417
Date published:
DOI:
doi:10.1038/nature12633
Received
Accepted
Published online

The appropriate timing of flowering is crucial for plant reproductive success. It is therefore not surprising that intricate genetic networks have evolved to perceive and integrate both endogenous and environmental signals, such as carbohydrate and hormonal status, photoperiod and temperature1, 2. In contrast to our detailed understanding of the vernalization pathway, little is known about how flowering time is controlled in response to changes in the ambient growth temperature. In Arabidopsis thaliana, the MADS-box transcription factor genes FLOWERING LOCUS M (FLM) and SHORT VEGETATIVE PHASE (SVP) have key roles in this process3, 4. FLM is subject to temperature-dependent alternative splicing3. Here we report that the two main FLM protein splice variants, FLM-β and FLM-δ, compete for interaction with the floral repressor SVP. The SVP–FLM-β complex is predominately formed at low temperatures and prevents precocious flowering. By contrast, the competing SVP–FLM-δ complex is impaired in DNA binding and acts as a dominant-negative activator of flowering at higher temperatures. Our results show a new mechanism that controls the timing of the floral transition in response to changes in ambient temperature. A better understanding of how temperature controls the molecular mechanisms of flowering will be important to cope with current changes in global climate5, 6.

At a glance

Figures

  1. Temperature-dependent expression of FLM-[bgr], FLM-[dgr] and SVP.
    Figure 1: Temperature-dependent expression of FLM-β, FLM-δ and SVP.

    a, FLM locus, including exons (boxes) and introns (lines). Primers used for qRT–PCR are indicated. bd, Relative (rel.) expression of FLM-β (light grey) and FLM-δ (dark grey) in different tissues in 10-day-old Col-0 seedlings grown at 23°C long-day (b), in whole seedlings at 16°C, 23°C and 27°C long-day (c) and days 1 and 5 after a shift between temperatures (d). The ratio of FLM-β/FLM-δ expression is shown in black. e, SVP expression in 10-day-old Col-0 seedlings grown at 16°C, 23°C and 27°C. Error bars denote s.d. of three biological replicates with three technical repetitions each.

  2. FLM-SVP protein-protein interactions and DNA binding assays.
    Figure 2: FLM-SVP protein–protein interactions and DNA binding assays.

    a, Yeast two-hybrid. AD, activation domain; BD, DNA-binding domain. b, BiFC FLM-β, FLM-δ and SVP interaction assays. C-citr., C-terminal half of haemagglutinin (HA)-tagged mCitrine; N-citr., N-terminal half of Myc-tagged mCitrine. c, d, Local enrichment of GFP-tagged FLM, iFLM-β and iFLM-δ bound to the SEP3 (c) and SOC1 (d) regulatory regions assayed by ChIP-seq. Each panel shows a 6-kilobase (kb) window. eg, EMSA competition assays using a SEP3 promoter probe containing two CArG motifs. Lanes 1 and 2 correspond to ‘no protein’ and ‘shuffled-SVP’ controls, respectively. Increasing concentrations of FLM-β (e) and FLM-δ (f) were added to a constant amount of SVP. g, Titration of FLM-δ to constant amounts of SVP and FLM-β. Orange and blue ellipses represent SVP and FLM-β proteins, respectively. h, Relative enrichment of binding of iFLM-β-GFP to the promoters of SOC1 (open bars) and SEP3 (filled bars) in iFLM-β-GFP × 35S:FLM-δ F1 and control plants. i, Expression of SOC1 and SEP3 in iFLM-β-GFP × 35S:FLM-δ F1 (light grey) and control (dark grey) plants. Error bars denote the s.d. of three biological replicates with three technical repetitions each. j, Flowering time of the F1 plants. Rosette and cauline leaf number are represented in dark and light grey.

  3. FLM and SVP are interdependent and regulate flowering time and flower morphology.
    Figure 3: FLM and SVP are interdependent and regulate flowering time and flower morphology.

    a, flm-3 suppresses the green and leaf-like sepal and petal phenotype of 35S:SVP flowers. b, Flowering time of flm-3 and svp-32 single and double mutants, and FLM-β and SVP mis-expression lines in wild-type (Col-0), svp-32 or flm-3. c, 35S:FLM-δ suppresses the 35S:SVP flower phenotype. d, 35S:FLM-β enhances the late flowering of 35S:SVP, whereas 35S:FLM-δ has the opposite effect. Plants were grown at 23°C (a, b) and 16°C (c, d). Rosette and cauline leaf number are represented in dark and light grey, respectively.

  4. Analysis of FLM splice variant expression in Col-0.
    Extended Data Fig. 1: Analysis of FLM splice variant expression in Col-0.

    a, Graphic representation of the FLM-α, FLM-β, FLM-γ and FLM-δ transcripts, including exons (boxes) and introns (lines). Primers used for FLM-α (F1-R1), FLM-β (F1-R2), FLM-γ (F2-R1) and FLM-δ (F2-R2) amplification are shown. b, Semi-quantitative RT–PCR of FLM splice variants in Col-0 cDNA at different temperatures, using plasmids for each splice variant as controls (lanes 1–4).

  5. Distribution of flowering time of independent transgenic T1 lines established in this study.
    Extended Data Fig. 2: Distribution of flowering time of independent transgenic T1 lines established in this study.

    ad, Flowering time of 35S:FLM-β and 35S:FLM-δ in Col-0 and flm-3 mutant background (a), flm-3 pFLM:gFLM and flm-3 pFLM:gFLM-GFP (b), flm-3 pFLM:iFLM-β, flm-3 pFLM:iFLM-δ, flm-3 pFLM:iFLM-β-GFP and flm-3 pFLM:iFLM-δ-GFP (c) and 35S:FLM-β-VP16 (d), grown under 16°C, long-day are shown. Shaded areas mark the median and the 25% and 75% percentile of flowering time for a given genotype.

  6. Analysis of FLM co-suppression in 35S:FLM-[dgr].
    Extended Data Fig. 3: Analysis of FLM co-suppression in 35S:FLM-δ.

    a, Analysis of FLM-β, FLM-δ, MAF2-5, FLC and SVP in Col-0 control and 35S:FLM-δ #4. All genes except FLM-δ and MAF4 were expressed at similar levels in Col-0 and FLM-δ overexpression line. b, Flowering time of maf-4 (SALK_028506) is similar to that of Col-0 control plants, indicating that the MAF4 downregulation observed in a cannot explain the early flowering phenotype of the 35S:FLM-δ line. ce, Flowering time (c) and expression (d) of FLM-β, and expression of FLM-δ (e), as determined by qRT–PCR analysis in F1 populations from crosses between 35S:FLM-β and 35S:FLM-δ plants in both Col-0 and flm-3 backgrounds. d, FLM-β expression is not co-suppressed in response to the FLM-δ misexpression (e) in both Col-0 and flm-3 backgrounds. Rosette and cauline leaf numbers after bolting are represented in dark and light grey, respectively, in b and c. Error bars denote the s.d. of three biological replicates with three technical repetitions each in a, d and e.

  7. BiFC competition experiment.
    Extended Data Fig. 4: BiFC competition experiment.

    ah, Microscope images (ad) and quantification of mCitrine-positive nuclei (eh). Increasing amounts (A600nm; bottom) of an untagged version of one of the interactors tested were included in the assay. The number of BiFC-positive nuclei decreases with increasing amounts of the specific competitor: FLM-β–FLM-β homodimerization (a, e) and FLM-δ–FLM-β (b, f), SVP–FLM-β (c, g) and FLM-δ–SVP (d, h) heterodimerization.

  8. Flowering time of the agl74N T-DNA mutant.
    Extended Data Fig. 5: Flowering time of the agl74N T-DNA mutant.

    Flowering time of Col-0, flm-3, and a homozygous agl74N T-DNA insertion line (SALK_016446).

  9. Graphic representation of iFLM-[bgr]/[dgr]-(GFP) constructs, gBrowse traces of mapped ChIP-seq reads and validation of FLM targets.
    Extended Data Fig. 6: Graphic representation of iFLM-β/δ-(GFP) constructs, gBrowse traces of mapped ChIP-seq reads and validation of FLM targets.

    a, iFLM-β/δ-(GFP) constructs representation including exons (boxes), introns included (black flat line) and introns missing (grey lines). Dashed boxes indicate presence only in the mGFP6-tagged constructs. be, Local enrichment of FLM, iFLM-β and iFLM-δ binding in ATC (b), RVE2 (c), SHP2 (d) and TEM2 (e). Chromosomal position (TAIR10) and models of the genes close to the peaks are given at the top of the panels. Each panel shows a 5-kb window. Forward reads are mapped above each line and reverse reads below. fh, qRT–PCR expression analysis of SEP3 (f), SOC1 (g) and TEM2 (h) in flm-3 mutant, Col-0 wild-type and a 35S:FLM-β transgenic line show how increasing levels of FLM-β downregulate SEP3 and SOC1 expression, but induce TEM2. Error bars in fh denote the s.d. of three biological replicates with three technical repetitions each.

  10. Venn diagram showing the number and overlap of FLM and SVP targets.
    Extended Data Fig. 7: Venn diagram showing the number and overlap of FLM and SVP targets.

    a, Overlap of loci bound in gFLM-GFP and iFLM-β-GFP ChIP-seq experiments with a false discovery rate (FDR)<0.1 in all biological replicates. At this FDR, the high quality iFLM-β-GFP data set identifies 460 targets that are missing from the gFLM-GFP data set, which includes a replicate (#2) that contains substantial fewer uniquely mappable reads than the other replicates (see Supplementary Table 2). b, Overlap of loci bound in gFLM-GFP and iFLM-β-GFP ChIP-seq (FDR <0.1) and SVP (FDR <0.05) ChIP-chip assays21.

  11. EMSA assays and FLM-[bgr]/SVP BiFC competition experiment.
    Extended Data Fig. 8: EMSA assays and FLM-β/SVP BiFC competition experiment.

    ac, EMSA assay with three sequences identified as binding-sites for SVP21 and FLM (this work) by ChIP-chip and ChIP-seq, respectively. SEP3 (a), SOC1 (b) and ATC (c) promoter probes that include two (a, b) or one (c) CArG motif(s) were used in EMSA. Different order complexes are represented by black arrowheads and asterisks for homo- or heterotetramers, respectively, and with grey arrowheads and asterisks for homo- or heterodimers, respectively. Orange and blue ellipses represent SVP and FLM-β proteins, respectively. d, e, Microscope images (d) and quantification (e) of mCitrine-positive nuclei. Increasing amounts (A600nm; bottom) of untagged 35S:FLM-δ were added to FLM-β and SVP mCitrine-tagged vectors. A reduction in the number of BiFC-positive nuclei is observed with increasing amounts of competitor.

  12. Models of SVP-FLM complex function.
    Extended Data Fig. 9: Models of SVP–FLM complex function.

    a, Temperature-dependent FLM splicing and genetic interactions of SVP–FLM-β heterocomplex in the flowering pathway. Strong binding of FLM to FT was observed in only one ChIP-seq replicate. Hence we propose that FLM–SVP downregulates FT expression in leaves indirectly through the induction of floral repressors transcription factors such as TEM2 and the AP2-like TOE3. The FLM–SVP complex contributes to the repression of floral transition by directly downregulating SOC1 and SEP3 expression, where SOC1 is a major floral activator. Arrows and block lines denote activation and repression, respectively. Dotted lines indicate a putative direct regulation. Rounded rectangles indicate proteins. b, Model of the temperature-dependent SVP–FLM complex function. Although SVP expression level is constant, FLM-β and FLM-δ levels are regulated in an antagonistic manner, with the former being the prevalent protein at low temperature and the latter dominating at high temperatures. At low temperatures SVP and FLM-β can interact, forming both homo- or heterocomplexes. The SVP-containing complexes are able to bind to the CArG boxes in the cis elements of important flowering related genes such as SEP3, SOC1, ATC, TEM2 and TOE3 and regulate their expression. When temperature increases, alternative splicing of FLM occurs, making FLM-δ the predominant splice variant. FLM-δ proteins compete with the remaining FLM-β and SVP proteins for complex formation. This results in the formation of non-functional SVP–FLM-δ complexes, which are impaired in their DNA-binding capability. The temperature-dependent splicing regulation of FLM occurs within 24h, allowing the plant to quickly sense and respond to changes in ambient temperature, ensuring the switch between the non-flowering and flowering phase of development.

Tables

  1. Mutants and transgenic lines used in this study.
    Extended Data Table 1: Mutants and transgenic lines used in this study.

Accession codes

Referenced accessions

Gene Expression Omnibus

References

  1. Srikanth, A. & Schmid, M. Regulation of flowering time: all roads lead to Rome. Cell. Mol. Life Sci. 68, 20132037 (2011)
  2. Posé, D., Yant, L. & Schmid, M. The end of innocence: flowering networks explode in complexity. Curr. Opin. Plant Biol. 15, 4550 (2012)
  3. Balasubramanian, S., Sureshkumar, S., Lempe, J. & Weigel, D. Potent induction of Arabidopsis thaliana flowering by elevated growth temperature. PLoS Genet. 2, e106 (2006)
  4. Lee, J. H. et al. Role of SVP in the control of flowering time by ambient temperature in Arabidopsis. Genes Dev. 21, 397402 (2007)
  5. Marcott, S. A., Shakun, J. D., Clark, P. U. & Mix, A. C. A reconstruction of regional and global temperature for the past 11,300 years. Science 339, 11981201 (2013)
  6. Mann, M. E. et al. Proxy-based reconstructions of hemispheric and global surface temperature variations over the past two millennia. Proc. Natl Acad. Sci. USA 105, 1325213257 (2008)
  7. Chouard, P. Vernalization and its relations to dormancy. Annu. Rev. Plant Physiol. Plant Mol. Biol. 11, 191238 (1960)
  8. Sheldon, C. C. et al. The FLF MADS box gene: a repressor of flowering in Arabidopsis regulated by vernalization and methylation. Plant Cell 11, 445458 (1999)
  9. Michaels, S. D. & Amasino, R. M. FLOWERING LOCUS C encodes a novel MADS domain protein that acts as a repressor of flowering. Plant Cell 11, 949956 (1999)
  10. Kumar, S. V. & Wigge, P. A. H2A.Z-containing nucleosomes mediate the thermosensory response in Arabidopsis. Cell 140, 136147 (2010)
  11. Kumar, S. V. et al. Transcription factor PIF4 controls the thermosensory activation of flowering. Nature 484, 242245 (2012)
  12. Franklin, K. A. et al. Phytochrome-interacting factor 4 (PIF4) regulates auxin biosynthesis at high temperature. Proc. Natl Acad. Sci. USA 108, 2023120235 (2011)
  13. Rosloski, S. M. et al. Functional analysis of splice variant expression of MADS AFFECTING FLOWERING 2 of Arabidopsis thaliana. Plant Mol. Biol. 81, 5769 (2013)
  14. Gu, X. et al. Arabidopsis FLC clade members form flowering-repressor complexes coordinating responses to endogenous and environmental cues. Nature Commun. 4, 1947 (2013)
  15. Werner, J. D. et al. Quantitative trait locus mapping and DNA array hybridization indentify an FLM deletion as a causefor natural flowering-time variation. Proc. Natl Acad. Sci. USA 102, 24602465 (2005)
  16. Méndez-Vigo, B., Martinez-Zapater, J. M. & Alonso-Blanco, C. The flowering repressor SVP underlies a novel Arabidopsis thaliana QTL interacting with the genetic background. PLoS Genet. 9, e1003289 (2013)
  17. Scortecci, K., Michaels, S. D. & Amasino, R. M. Genetic interactions between FLM and other flowering-time genes in Arabidopsis thaliana. Plant Mol. Biol. 52, 915922 (2003)
  18. Scortecci, K. C., Michaels, S. D. & Amasino, R. M. Identification of a MADS-box gene, FLOWERING LOCUS M, that represses flowering. Plant J. 26, 229236 (2001)
  19. Jiao, Y. & Meyerowitz, E. M. Cell-type specific analysis of translating RNAs in developing flowers reveals new levels of control. Mol. Syst. Biol. 6, 419 (2010)
  20. Riechmann, J. L., Krizek, B. A. & Meyerowitz, E. M. Dimerization specificity of Arabidopsis MADS domain homeotic proteins APETALA1, APETALA3, PISTILLATA, and AGAMOUS. Proc. Natl Acad. Sci. USA 93, 47934798 (1996)
  21. Tao, Z. et al. Genome-wide identification of SOC1 and SVP targets during the floral transition in Arabidopsis. Plant J. 70, 549561 (2012)
  22. James, A. B. et al. Alternative splicing mediates responses of the Arabidopsis circadian clock to temperatures changes. Plant Cell 24, 961981 (2012)
  23. Wang, X. et al. SKIP is a component of the spliceosome linking alternative splicing and the circadian clock in Arabidopsis. Plant Cell 24, 32783295 (2012)
  24. Jones, M. A. et al. Mutation of Arabidopsis spliceosomal timekeeper locus1 causes circadian clock defects. Plant Cell 24, 40664082 (2012)
  25. Curtis, M. D. & Grossniklaus, U. A gateway cloning vector set for high-throughput functional analysis of genes in planta. Plant Physiol. 133, 462469 (2003)
  26. Clough, S. J. & Bent, A. F. Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J. 16, 735743 (1998)
  27. Immink, R. G. et al. Characterization of SOC1’s central role in flowering by the identification of its upstream and downstream regulators. Plant Physiol. 160, 433449 (2012)
  28. Yant, L. et al. Orchestration of the floral transition and floral development in Arabidopsis by the bifunctional transcription Factor APETALA2. Plant Cell 22, 21562170 (2010)
  29. Moyroud, E. et al. Prediction of Regulatory Interactions from Genome Sequences Using a Biophysical Model for the Arabidopsis LEAFY Transcription Factor. Plant Cell 23, 12931306 (2011)
  30. James, P., Halladay, J. & Craig, E. A. Genomic libraries and a host strain designed for highly efficient two-hybrid selection in yeast. Genetics 144, 14251436 (1996)
  31. de Folter, S. et al. Comprehensive interaction map of the Arabidopsis MADS Box transcription factors. Plant Cell 17, 14241433 (2005)
  32. Severing, E. I. et al. Predicting the impact of alternative splicing on plant MADS domain protein function. PLoS ONE 7, e30524 (2012)
  33. van Dijk, A. D. et al. Sequence motifs in MADS transcription factors responsible for specificity and diversification of protein-protein interaction. PLOS Comput. Biol. 6, e1001017 (2010)
  34. de Felippes, F. F. & Weigel, D. Transient assays for the analysis of miRNA processing and function. Methods Mol. Biol. 592, 255264 (2010)
  35. Masiero, S. et al. INCOMPOSITA: a MADS-box gene controlling prophyll development and floral meristem identity in Antirrhinum. Development 131, 59815990 (2004)

Download references

Author information

Affiliations

  1. Max Planck Institute for Developmental Biology, Department of Molecular Biology, Spemannstr. 35, 72076 Tübingen, Germany

    • David Posé,
    • Felix Ott,
    • Levi Yant,
    • Johannes Mathieu &
    • Markus Schmid
  2. Plant Research International, Bioscience, Droevendaalsesteeg 1, 6708 PB Wageningen, The Netherlands

    • Leonie Verhage,
    • Gerco C. Angenent &
    • Richard G. H. Immink
  3. Laboratory of Molecular Biology, Wageningen University, 6708 PB Wageningen, The Netherlands

    • Leonie Verhage &
    • Gerco C. Angenent
  4. Present addresses: Instituto de Hortofruticultura Subtropical y Mediterránea, Universidad de Málaga–Consejo Superior de Investigaciones Científicas, Departamento de Biología Molecular y Bioquímica, Facultad de Ciencias, Universidad de Málaga, 29071 Málaga, Spain (D.P.); Department of Organismic and Evolutionary Biology, Harvard University, 16 Divinity Avenue, Cambridge, Massachusetts 02138, USA (L.Y.); Boyce Thompson Institute for Plant Research, Tower Road, Ithaca, New York 14853-1801, USA (J.M.).

    • David Posé,
    • Levi Yant &
    • Johannes Mathieu

Contributions

D.P. performed most experiments, L.V. performed EMSA assays, F.O. analysed ChIP-seq data, L.Y. and J.M. generated 35S:FLM-β/35S:FLM-δ constructs, L.Y. generated and phenotyped F1 Col-0 35S:FLM-β/35S:FLM-δ transgenic plants, G.C.A. contributed to the discussion, and D.P., R.G.H.I., L.Y. and M.S. supervised the work and wrote the manuscript.

Competing financial interests

The authors declare no competing financial interests.

Corresponding author

Correspondence to:

ChIP-seq data have been deposited with NCBI Gene Expression Omnibus under accession number GSE48082.

Author details

Extended data figures and tables

Extended Data Figures

  1. Extended Data Figure 1: Analysis of FLM splice variant expression in Col-0. (148 KB)

    a, Graphic representation of the FLM-α, FLM-β, FLM-γ and FLM-δ transcripts, including exons (boxes) and introns (lines). Primers used for FLM-α (F1-R1), FLM-β (F1-R2), FLM-γ (F2-R1) and FLM-δ (F2-R2) amplification are shown. b, Semi-quantitative RT–PCR of FLM splice variants in Col-0 cDNA at different temperatures, using plasmids for each splice variant as controls (lanes 1–4).

  2. Extended Data Figure 2: Distribution of flowering time of independent transgenic T1 lines established in this study. (185 KB)

    ad, Flowering time of 35S:FLM-β and 35S:FLM-δ in Col-0 and flm-3 mutant background (a), flm-3 pFLM:gFLM and flm-3 pFLM:gFLM-GFP (b), flm-3 pFLM:iFLM-β, flm-3 pFLM:iFLM-δ, flm-3 pFLM:iFLM-β-GFP and flm-3 pFLM:iFLM-δ-GFP (c) and 35S:FLM-β-VP16 (d), grown under 16°C, long-day are shown. Shaded areas mark the median and the 25% and 75% percentile of flowering time for a given genotype.

  3. Extended Data Figure 3: Analysis of FLM co-suppression in 35S:FLM-δ. (265 KB)

    a, Analysis of FLM-β, FLM-δ, MAF2-5, FLC and SVP in Col-0 control and 35S:FLM-δ #4. All genes except FLM-δ and MAF4 were expressed at similar levels in Col-0 and FLM-δ overexpression line. b, Flowering time of maf-4 (SALK_028506) is similar to that of Col-0 control plants, indicating that the MAF4 downregulation observed in a cannot explain the early flowering phenotype of the 35S:FLM-δ line. ce, Flowering time (c) and expression (d) of FLM-β, and expression of FLM-δ (e), as determined by qRT–PCR analysis in F1 populations from crosses between 35S:FLM-β and 35S:FLM-δ plants in both Col-0 and flm-3 backgrounds. d, FLM-β expression is not co-suppressed in response to the FLM-δ misexpression (e) in both Col-0 and flm-3 backgrounds. Rosette and cauline leaf numbers after bolting are represented in dark and light grey, respectively, in b and c. Error bars denote the s.d. of three biological replicates with three technical repetitions each in a, d and e.

  4. Extended Data Figure 4: BiFC competition experiment. (768 KB)

    ah, Microscope images (ad) and quantification of mCitrine-positive nuclei (eh). Increasing amounts (A600nm; bottom) of an untagged version of one of the interactors tested were included in the assay. The number of BiFC-positive nuclei decreases with increasing amounts of the specific competitor: FLM-β–FLM-β homodimerization (a, e) and FLM-δ–FLM-β (b, f), SVP–FLM-β (c, g) and FLM-δ–SVP (d, h) heterodimerization.

  5. Extended Data Figure 5: Flowering time of the agl74N T-DNA mutant. (33 KB)

    Flowering time of Col-0, flm-3, and a homozygous agl74N T-DNA insertion line (SALK_016446).

  6. Extended Data Figure 6: Graphic representation of iFLM-β/δ-(GFP) constructs, gBrowse traces of mapped ChIP-seq reads and validation of FLM targets. (251 KB)

    a, iFLM-β/δ-(GFP) constructs representation including exons (boxes), introns included (black flat line) and introns missing (grey lines). Dashed boxes indicate presence only in the mGFP6-tagged constructs. be, Local enrichment of FLM, iFLM-β and iFLM-δ binding in ATC (b), RVE2 (c), SHP2 (d) and TEM2 (e). Chromosomal position (TAIR10) and models of the genes close to the peaks are given at the top of the panels. Each panel shows a 5-kb window. Forward reads are mapped above each line and reverse reads below. fh, qRT–PCR expression analysis of SEP3 (f), SOC1 (g) and TEM2 (h) in flm-3 mutant, Col-0 wild-type and a 35S:FLM-β transgenic line show how increasing levels of FLM-β downregulate SEP3 and SOC1 expression, but induce TEM2. Error bars in fh denote the s.d. of three biological replicates with three technical repetitions each.

  7. Extended Data Figure 7: Venn diagram showing the number and overlap of FLM and SVP targets. (121 KB)

    a, Overlap of loci bound in gFLM-GFP and iFLM-β-GFP ChIP-seq experiments with a false discovery rate (FDR)<0.1 in all biological replicates. At this FDR, the high quality iFLM-β-GFP data set identifies 460 targets that are missing from the gFLM-GFP data set, which includes a replicate (#2) that contains substantial fewer uniquely mappable reads than the other replicates (see Supplementary Table 2). b, Overlap of loci bound in gFLM-GFP and iFLM-β-GFP ChIP-seq (FDR <0.1) and SVP (FDR <0.05) ChIP-chip assays21.

  8. Extended Data Figure 8: EMSA assays and FLM-β/SVP BiFC competition experiment. (211 KB)

    ac, EMSA assay with three sequences identified as binding-sites for SVP21 and FLM (this work) by ChIP-chip and ChIP-seq, respectively. SEP3 (a), SOC1 (b) and ATC (c) promoter probes that include two (a, b) or one (c) CArG motif(s) were used in EMSA. Different order complexes are represented by black arrowheads and asterisks for homo- or heterotetramers, respectively, and with grey arrowheads and asterisks for homo- or heterodimers, respectively. Orange and blue ellipses represent SVP and FLM-β proteins, respectively. d, e, Microscope images (d) and quantification (e) of mCitrine-positive nuclei. Increasing amounts (A600nm; bottom) of untagged 35S:FLM-δ were added to FLM-β and SVP mCitrine-tagged vectors. A reduction in the number of BiFC-positive nuclei is observed with increasing amounts of competitor.

  9. Extended Data Figure 9: Models of SVP–FLM complex function. (342 KB)

    a, Temperature-dependent FLM splicing and genetic interactions of SVP–FLM-β heterocomplex in the flowering pathway. Strong binding of FLM to FT was observed in only one ChIP-seq replicate. Hence we propose that FLM–SVP downregulates FT expression in leaves indirectly through the induction of floral repressors transcription factors such as TEM2 and the AP2-like TOE3. The FLM–SVP complex contributes to the repression of floral transition by directly downregulating SOC1 and SEP3 expression, where SOC1 is a major floral activator. Arrows and block lines denote activation and repression, respectively. Dotted lines indicate a putative direct regulation. Rounded rectangles indicate proteins. b, Model of the temperature-dependent SVP–FLM complex function. Although SVP expression level is constant, FLM-β and FLM-δ levels are regulated in an antagonistic manner, with the former being the prevalent protein at low temperature and the latter dominating at high temperatures. At low temperatures SVP and FLM-β can interact, forming both homo- or heterocomplexes. The SVP-containing complexes are able to bind to the CArG boxes in the cis elements of important flowering related genes such as SEP3, SOC1, ATC, TEM2 and TOE3 and regulate their expression. When temperature increases, alternative splicing of FLM occurs, making FLM-δ the predominant splice variant. FLM-δ proteins compete with the remaining FLM-β and SVP proteins for complex formation. This results in the formation of non-functional SVP–FLM-δ complexes, which are impaired in their DNA-binding capability. The temperature-dependent splicing regulation of FLM occurs within 24h, allowing the plant to quickly sense and respond to changes in ambient temperature, ensuring the switch between the non-flowering and flowering phase of development.

Extended Data Tables

  1. Extended Data Table 1: Mutants and transgenic lines used in this study. (191 KB)

Supplementary information

Excel files

  1. Supplementary Table 1 (51 KB)

    This file contains a yeast two-hybrid analysis of FLM-β and FLM-δ interactions with a collection of Arabidopsis MADS domain transcription factors.

  2. Supplementary Table 2 (91 KB)

    This table contains the read numbers for pFLM:gFLM-GFP.

  3. Supplementary Table 3 (286 KB)

    This table contains the read numbers for pFLM:iFLM-β-GFP.

  4. Supplementary Table 4 (178 KB)

    This file contains 232 common targets. Please note that pFLM:gFLM-GFP and pFLM:iFLM-β-GFP are applicable to this study.

  5. Supplementary Table 5 (37 KB)

    This file contains the Oligonucleotides used in this study.

Additional data